You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@shiro.apache.org by "Matt Bishop (JIRA)" <ji...@apache.org> on 2016/03/21 20:10:25 UTC

[jira] [Updated] (SHIRO-564) WildcardPermission case-insensitive makes parts collections twice

     [ https://issues.apache.org/jira/browse/SHIRO-564?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Matt Bishop updated SHIRO-564:
------------------------------
    Description: 
WildcardPermissions have a caseSensitive flag that instructs the permission to ignore case when comparing permissions. It's implementation of this feature is kinda wasteful because it first parses the permission string into a Set<String> part, THEN checks the flag and then rebuilds the part Set.

It might be faster to check the flag first, lowercase the permission string, then parse the parts. This would save creating a whole bunch of extra HashSets.

  was:
WildcardPermissions have a caseSensitive flag that instructs the permission to ignore case when comparing permissions. It's implementation of this feature is kinda wasteful because it first parses the permission string into List<Set<String>> parts, THEN checks the flag and then rebuilds the parts collections.

It might be faster to check the flag first, lowercase the permission string, then parse the parts. This would save creating a whole bunch of extra HashSets and ArrayLists.


> WildcardPermission case-insensitive makes parts collections twice
> -----------------------------------------------------------------
>
>                 Key: SHIRO-564
>                 URL: https://issues.apache.org/jira/browse/SHIRO-564
>             Project: Shiro
>          Issue Type: Improvement
>          Components: Authorization (access control) 
>    Affects Versions: 1.2.4
>         Environment: Any
>            Reporter: Matt Bishop
>            Priority: Trivial
>
> WildcardPermissions have a caseSensitive flag that instructs the permission to ignore case when comparing permissions. It's implementation of this feature is kinda wasteful because it first parses the permission string into a Set<String> part, THEN checks the flag and then rebuilds the part Set.
> It might be faster to check the flag first, lowercase the permission string, then parse the parts. This would save creating a whole bunch of extra HashSets.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)